On Sat, 2010-02-13 at 21:20 -0500, Kyle Lil wrote: > > From: [email protected]
> OK, Great. So, enabling the debug messages as you described results in > a ton of these messages: > > > [ 1312.000257] cx18-0: warning: failed to be awakened upon RPU > acknowledgment sending CX18_CPU_DE_SET_MDL; timed out waiting 10 msecs > [ 1312.500011] cx18-0: warning: failed to be awakened upon RPU > acknowledgment sending CX18_CPU_DE_SET_MDL; timed out waiting 10 msecs > [ 1313.000017] cx18-0: warning: failed to be awakened upon RPU > acknowledgment sending CX18_CPU_DE_SET_MDL; timed out waiting 10 msecs > [ 1315.080014] cx18-0: warning: failed to be awakened upon RPU > acknowledgment sending CX18_CPU_DE_SET_MDL; timed out waiting 10 msecs > [ 1315.310061] cx18-0: warning: failed to be awakened upon RPU > acknowledgment sending CX18_CPU_DE_SET_MDL; timed out waiting 10 msecs > [ 1316.330025] cx18-0: warning: failed to be awakened upon RPU > acknowledgment sending CX18_CPU_DE_SET_MDL; timed out waiting 10 msecs > [ 1316.870079] cx18-0: warning: failed to be awakened upon RPU > acknowledgment sending CX18_CPU_DE_SET_MDL; timed out waiting 10 msecs Those aren't so bad. These happen when we give a buffer bad to the CX23418 and it doesn't respond *right away*. The cx18-0-out/n work handler will go to sleep waiting for a wake-up from the CX23418 via an IRQ. Usually the ACK IRQ happens well within 10 msecs, but the Linux scheduler doesn't get around to waking up the cx18-0-out/n work handler, so you get this message instead when a timer ends up waking it up. Short story: no big deal as far as the CX23418 is concerned. On the other hand, your system load or interrupt handling latency may be high. > and an occasional message like this: > > > [ 1311.480015] cx18-0: warning: sending CX18_CPU_DE_SET_MDL timed out > waiting 10 msecs for RPU acknowledgement This is an instance when the CX23418 firmware actually didn't ACK us within 10 msecs of us sending an empty buffer back. This means the firmware was very busy at the moment, or was ignoring us because it didn't make sense for us to be giving it a buffer at that time (i.e. we were stopping the streaming). This isn't a big deal unless it happens often. > Executing mplayer as you described now attempts to fill the cache and > play video, but is taking forever to fill the cache. To get 0.1% (8192 > bytes) takes 13 seconds. It hasn't yet completely filled the cache, > but it has given a couple of error messages saying > "dvb_streaming_read, attempt N. 6 failed with errno 0 when reading x > bytes". It is now at about 10% That's certainly not what should happen normally. With a good stream (or good signal?), it should get up to 19% cache fill and start playing within about 5 seconds. Like so: $ time mplayer dvb://WTTG\ DT -cache 8192 MPlayer SVN-r28461-4.3.2 (C) 2000-2009 MPlayer Team CPU: AMD Athlon(tm) 64 X2 Dual Core Processor 4200+ (Family: 15, Model: 43, Stepping: 1) mplayer: could not open config files /home/andy/.lircrc and /etc/lircrc mplayer: No such file or directory Failed to read LIRC config file ~/.lircrc. Playing dvb://WTTG DT. dvb_tune Freq: 605028615 Cache fill: 19.14% (1605632 bytes) TS file format detected. VIDEO MPEG2(pid=49) AUDIO A52(pid=52) NO SUBS (yet)! PROGRAM N. 0 VIDEO: MPEG2 1280x720 (aspect 3) 59.940 fps 19000.0 kbps (2375.0 kbyte/s) ========================================================================== Opening video decoder: [mpegpes] MPEG 1/2 Video passthrough VDec: vo config request - 1280 x 720 (preferred colorspace: Mpeg PES) Could not find matching colorspace - retrying with -vf scale... Opening video filter: [scale] The selected video_out device is incompatible with this codec. Try appending the scale filter to your filter list, e.g. -vf spp,scale instead of -vf spp. VDecoder init failed :( Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family Selected video codec: [ffmpeg2] vfm: ffmpeg (FFmpeg MPEG-2) ========================================================================== ========================================================================== Opening audio decoder: [liba52] AC3 decoding with liba52 Using SSE optimized IMDCT transform Using MMX optimized resampler AUDIO: 48000 Hz, 2 ch, s16le, 448.0 kbit/29.17% (ratio: 56000->192000) Selected audio codec: [a52] afm: liba52 (AC3-liba52) ========================================================================== AO: [pulse] 48000Hz 2ch s16le (2 bytes per sample) Starting playback... VDec: vo config request - 1280 x 720 (preferred colorspace: Planar YV12) VDec: using Planar YV12 as output csp (no 0) Movie-Aspect is 1.78:1 - prescaling to correct movie aspect. VO: [xv] 1280x720 => 1280x720 Planar YV12 A:49184.3 V:49184.7 A-V: -0.335 ct: -0.065 40/ 40 126% 6% 5.1% 4 0 14% Exiting... (Quit) real 0m4.303s user 0m0.604s sys 0m0.148s With a poor stream due to weak signal, I get something like this with ATSC: $ time mplayer dvb://WJLA-HD -cache 8192 MPlayer SVN-r28461-4.3.2 (C) 2000-2009 MPlayer Team CPU: AMD Athlon(tm) 64 X2 Dual Core Processor 4200+ (Family: 15, Model: 43, Stepping: 1) mplayer: could not open config files /home/andy/.lircrc and /etc/lircrc mplayer: No such file or directory Failed to read LIRC config file ~/.lircrc. Playing dvb://WJLA-HD. dvb_tune Freq: 177028615 Cache fill: 8.79% (737280 bytes) dvb_streaming_read, attempt N. 6 failed with errno 0 when reading 844 bytes Cache fill: 8.79% (737280 bytes) dvb_streaming_read, attempt N. 5 failed with errno 0 when reading 844 bytes Cache fill: 19.34% (1622016 bytes) TS file format detected. VIDEO MPEG2(pid=49) AUDIO A52(pid=52) NO SUBS (yet)! PROGRAM N. 0 VIDEO: MPEG2 1280x720 (aspect 3) 59.940 fps 17782.8 kbps (2222.8 kbyte/s) ========================================================================== Opening video decoder: [mpegpes] MPEG 1/2 Video passthrough VDec: vo config request - 1280 x 720 (preferred colorspace: Mpeg PES) Could not find matching colorspace - retrying with -vf scale... Opening video filter: [scale] The selected video_out device is incompatible with this codec. Try appending the scale filter to your filter list, e.g. -vf spp,scale instead of -vf spp. VDecoder init failed :( Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family Selected video codec: [ffmpeg2] vfm: ffmpeg (FFmpeg MPEG-2) ========================================================================== ========================================================================== Opening audio decoder: [liba52] AC3 decoding with liba52 Using SSE optimized IMDCT transform a52: CRC check failed! Using MMX optimized resampler AUDIO: 48000 Hz, 2 ch, s16le, 384.0 kbit/25.00% (ratio: 48000->192000) Selected audio codec: [a52] afm: liba52 (AC3-liba52) ========================================================================== AO: [pulse] 48000Hz 2ch s16le (2 bytes per sample) Starting playback... a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling VDec: vo config request - 1280 x 720 (preferred colorspace: Planar YV12) VDec: using Planar YV12 as output csp (no 0) Movie-Aspect is 1.78:1 - prescaling to correct movie aspect. VO: [xv] 1280x720 => 1280x720 Planar YV12 [mpeg2video @ 0xac4480]skipped MB in I frame at 65 5 [mpeg2video @ 0xac4480]Warning MVs not available [mpeg2video @ 0xac4480]concealing 1984 DC, 1984 AC, 1984 MV errors a52: CRC check failed! 1481.898 ct: 0.000 1/ 1 ??% ??% ??,?% 0 0 1% a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling [mpeg2video @ 0xac4480]concealing 2000 DC, 2000 AC, 2000 MV errors a52: CRC check failed! 2.479 ct: 0.002 2/ 2 ??% ??% ??,?% 0 0 0% a52: error at resampling a52: CRC check failed! [mpeg2video @ 0xac4480]concealing 2000 DC, 2000 AC, 2000 MV errors a52: CRC check failed! 2.496 ct: 0.003 3/ 3 ??% ??% ??,?% 0 0 0% a52: error at resampling a52: CRC check failed! a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling [mpeg2video @ 0xac4480]ac-tex damaged at 62 28 [mpeg2video @ 0xac4480]Warning MVs not available [mpeg2video @ 0xac4480]concealing 1352 DC, 1352 AC, 1352 MV errors a52: CRC check failed! 3.353 ct: 0.005 4/ 4 ??% ??% ??,?% 0 0 0% a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! 3.537 ct: 0.007 5/ 5 ??% ??% ??,?% 0 0 0% a52: error at resampling a52: CRC check failed! a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! 3.844 ct: 0.008 6/ 6 ??% ??% ??,?% 1 0 0% a52: error at resampling a52: CRC check failed! a52: error at resampling a52: CRC check failed! a52: error at resampling A:91484.4 V:91480.1 A-V: 4.286 ct: 0.010 7/ 7 ??% ??% ??,?% 2 0 0% Exiting... (Quit) real 0m8.538s user 0m0.318s sys 0m0.163s So mplayer is not very robust when dealing with streams with errors. It is very good at indicating to you that you have a stream with errors apparently. Windows might be able to better render smoothly through the errors. BTW, what does the Windows utility form Hauppauge show for signal strength? (I haven't used it, I just heard that Hauppauge provides one with its osftware.) The only other thing I can think of is that the cable signal could be slightly overamplified so it is overdriving the front end of the tuner and introducing these errors. You might want to put an attenuator in line to reduce the RF level a bit and see if that helps. If you can't find an in-line attenuator at a local electronics store, here's how to build a Bridged-T attenuator from resistors yourself: http://www.mail-archive.com/[email protected]/msg08735.html I guess you would want to try a "Gain" of about -3 dB first and then -10 dB if the problem still seems to happen. And of course there's always the checklist here to go through: http://www.ivtvdriver.org/index.php/Howto:Improve_signal_quality Regards, Andy _______________________________________________ ivtv-users mailing list [email protected] http://ivtvdriver.org/mailman/listinfo/ivtv-users
